Numbers reveal Trumbo's hidden value

Outfielder receives boost from metrics if he moves back to first base

January 6th, 2017

Despite hitting a Major League-leading 47 homers in 2016, remains unsigned, and it seems all but certain that he'll end up with a contract worth far less than the $75 million deal he was reportedly seeking at the Winter Meetings just a month ago. In one way, that reflects the new reality of baseball, because power-hitting sluggers who don't offer much value in getting on base or on defense simply don't get paid the way they used to, particularly ones who have the Qualifying Offer hanging over them as Trumbo does.
But what if there's a simple way to make Trumbo look more valuable? What if a team simply stopped putting him in a position to fail -- i.e., the outfield -- and let him play first base on a full-time basis, where he's been quietly valuable when he's had the opportunity over his career? What would his value look like then?
We can find out. Let's do a little math. Not too much, though. 

Last year, Trumbo was worth 2.2 Wins Above Replacement, per FanGraphs. A two-win player is considered to be about league average, and that placed him in a group of solid-but-not-elite players like Joe Panik, J.J. Hardy, and . That's partially because he was below-average at getting on base (his .316 OBP was below the .326 MLB average for non-pitchers) and on the bases, but largely because he was a negative on defense, and WAR attempts to estimate a player's entire value. Trumbo's one big plus was weighed down in other areas, so the net result was "average."
While current defensive metrics have their imperfections, this should be one time where the numbers and the eye test align pretty well. Of the 74 outfielders who had 750 innings last year, he was tied for 66th in Defensive Runs Savedand 69th in Ultimate Zone Rating. Going back to 2002, he's in the bottom 11th percentile in outfield UZR, and looking at WAR totals for 40-homer hitters since World War II, his 2016 season ranked 287th of 295
Compare him, for example, to , who played about a similar amount of innings in right field in 2016 and just slightly more outfield overall. As you can see in the rotating image below, using Statcast™ measurements on an outfielder's distance from the ball and the hang time he had in which to get there, Bourjos made nine catches considered "highlights" (the darkest red area) while Trumbo made none. Bourjos had 22 catches considered "tough," (the lighter red), while Trumbo had only seven.
Gif: Mark Trumbo and Peter Bourjos outfield catches graph
This is not meant to be controversial or derogatory; he's simply been asked to do something he's not strong at. He hasn't been placed in the best position to succeed.
That's happened repeatedly over his career, by the way. When he came up with the Angels, he was their primary first baseman in 2011 before arrived in 2012, pushing Trumbo to third and then the outfield. He'd see more time at first in 2013 with Pujols injured, but a 2014 trade to Arizona left him blocked at first base by . With Seattle in 2015, the Mariners largely used and at first, and finally in Baltimore in 2016, the O's had Chris Davis starting 152 games. If he returns to Baltimore, he'd probably again see most of his time in right field, especially since newly-acquiredSeth Smith is a lefty-swinging platoon bat.
That's a misuse of Trumbo's skills. Over his career, he's played roughly the same amount of time at first base (3,046 innings) as he has in the outfield (2,812 innings). According to UZR, he's been +15 runs at first in that time, and -21 runs in the outfield. If we put those numbers per 150 games, that's +6 at first (which ranks very well among his peers) and -10 in the outfield. These are big differences. Again, the numbers and eye test align well here.

So, here's our experiment: What would Trumbo's WAR total look like if we look back at his 2016 hitting line and tweak the math so he was playing a good first base, rather than a poor outfield? Rather than Panik, Hardy, and Grichuk, who would his comparables be?
Here's where the math comes in, but we'll keep it simple. There's really two things you need to know about how defensive numbers enter into WAR. The first is that different positions are assigned different difficulties -- shortstop, for example, is much harder to play than left field -- and first base is viewed as an easier position than right field, so there's a higher bar to clear. The second is that Trumbo has shown himself to be a much better first baseman than outfielder, so he should clear that bar and then some.
We'll take that knowledge and the 791 below-average innings that Trumbo played in the outfield in 2016, and instead credit him for a hypothetical 791 average innings at first base. We'll also go a step further and see what he'd look like with an above-average 791 innings at first base, using the baseline he himself has set at the position.

If Trumbo Played An Average 791 Innings At First Base In 2016…
...Trumbo's WAR goes from 2.1 to 2.7. What we've done here is to subtract 791 poor innings out of outfield play and replace them with 791 innings of league-average first base play, accounting for the fact that first base is considered "easier." Since league-average play at the "easiest" position isn't that valuable, this helps him just a bit (it doesn't change his limitations in getting on base or base-running, obviously), but suddenly his comparables become , Troy Tulowitzki and Davis -- much more impressive.
If Trumbo Played An Above-Average 791 Innings At First Base In 2016...
...Trumbo's WAR goes from 2.1 to 3.5. Of course, Trumbo has over 3,000 innings of first base play on his resume that's been better than average, and it's that number we've plugged into the equation here. If we turn him from a bad outfielder into a good first baseman based on his established ratings there, then suddenly Trumbo's 2016 WAR is in the same range as and . That's not the same thing as saying he's as good as those players, but we all know Harper struggled for months after a hot start, and this might be the range you'd expect such a power bat to live in.
And remember, we're only converting the 791 innings of outfield time he had, without separately including the 56 games he started at DH, which add no defensive value and which wouldn't exist if he joined a National League team. (The Rockies remain the obvious fit, even with in the fold, but the A's, who have been linked to Trumbo and could easily improve upon , or the Rangers, who lost and Mitch Moreland and have not yet finalized a rumored deal with Mike Napoli, also make sense.)
We're obviously making some assumptions here, and playing with the math. But the reason a lot of people don't like WAR is because it can say the top home-run hitter in the game was an average player. The problem, however, isn't with the metric. The problem is with the position Trumbo's been put in. A powerful hitter adding value at a position is always going to be more valuable than a powerful hitter subtracting value at a position, even with the positional adjustment included.
Had Trumbo been playing first base headed into his free agent year, he might have found a better marketplace. If Trumbo's next team doesn't force him to play the outfield, both sides will be far better off.